A 16-year-old male patient clinically diagnosed with LGMDD4 from a nonconsanguineous Han Chinese family was investigated (Fig. 1A). The mother and younger sister of the proband were also clinically diagnosed with LGMDD4 while the father was phenotypically normal.

A muscle biopsy was performed for the proband. Frozen sections of skeletal muscle were evaluated at The Reproductive Medicine Centre, The First Hospital of Lanzhou University (Lanzhou, China).

Excess moisture was first removed from the skeletal tissue sample simply by blotting thoroughly with a paper towel. Optimal cutting temperature compound was placed in the bottom of a shallow cryomold to provide a foundation for the section of muscle. The skeletal muscle section was carefully placed into the mold. Isopentane (2-methylbutane), which has a high thermal conductivity, does not form a vapor halo. Therefore, isopentane chilled with liquid nitrogen freezes muscle tissues more effectively and evenly than putting the tissues directly into liquid nitrogen. Therefore, prechilled isopentane was then chilled and maintained at -140˚C in a flask by adding liquid nitrogen. Then, the skeletal muscle section was kept in isopentane for 10-20 sec. Fresh frozen sections of skeletal muscle were stored at -80˚C. Then, cryo-sectioning of the frozen sections of skeletal muscles was performed and 10-µm sections were produced (20). These slides were processed further for staining.

Slides were incubated at room temperature with hematoxylin solution for 10 min and washed with distilled water. Next, the slides were stained with eosin solution for 5 min at room temperature, followed by treatment with ethanol [70% ethanol for 30 sec, 90% ethanol for 30 sec and absolute (100%) ethanol for 1 min] at room temperature and finally xylene for 3 min at room temperature (21). Then, the slides were air-dried, mounted with xylene and stored at room temperature. Digital microscopy (variation of a traditional optical/light microscope; total magnification, x200) was used to capture the H&E-stained images.

Firstly, TRIS buffer (0.05 M, pH 7.6), NADH solution, Nitro blue tetrazolium (NBT) solution and acetone de-staining solutions (30, 60 and 90%) were prepared. The NADH solution was gently thawed and 5 ml NBT solution was added to it. The slides were incubated in this solution for 30 min at 37˚C and washed three times with deionized water. The slides were washed again with acetone de-staining solutions first in increasing and then in decreasing concentrations. Finally, the slides were rinsed three times with deionized water and mounted with glycerol and phenol (22). A light microscope (total magnification, x200) was used to capture the images.

The frozen skeletal tissue block was transferred to a cryotome cryostat at -20˚C prior to sectioning and the temperature of the frozen skeletal muscle tissue block was allowed to equilibrate to the temperature of the cryotome cryostat. Sections of the frozen skeletal muscle tissue block of the desired thickness (3, 5 and 8 µm) were prepared. Then, the skeletal muscle tissue sections were placed onto glass slides suitable for immunohistochemistry. Next, these sections were dipped in distilled water and treated with 5% hydrogen peroxidase for endogenous peroxidase blocking and incubated for 30 min at room temperature. Then, these sections were incubated with BondTM Primary Antibody Diluent 0.5L (cat. no. AR9352; Leica Microsystems, Inc.) at room temperature for 30 min for non-specific background blocking. Immunohistochemical analysis of skeletal muscle samples was performed with four primary antibodies (dystrophin, α-sarcoglycan, β-sarcoglycan and caveolin-3) and Chromogenic Multiplex IHC for BOND RX/RX^m (Leica Microsystems, Inc.) was used as detection agent (23,24). Anti-Mouse UltraPolymer HRP (2MH-050: Anti-Mouse UltraPolymer HRP 50 ml; Cell IDx; Leica Microsystems, Inc.) was used to increase the sensitivity and signal amplification. The following primary antibodies were used: Dystrophin (dilution, 1:100; cat. no. ab275391; Abcam; 60 min at 25˚C), α-sarcoglycan (dilution, 1:1,000; cat. no. ab189254; Abcam; 60 min at 25˚C), β-sarcoglycan (dilution, 1:100; cat. no. ab135954; Abcam; 60 min at 25˚C) and caveolin-3 (dilution, 1:1,000; cat. no. ab289544; Abcam; 60 min at 25˚C). The following secondary antibody was used: Anti-Rabbit IgG H&L (HRP) (dilution, 1:2,000; cat. no. ab205718; Abcam; 10 h at 4˚C). A light microscope (total magnification, x200) was used to capture the images.

A blood sample was collected from the proband and genomic DNA was extracted (QIAamp DNA Blood Mini Kit; Qiagen GmbH) according to the manufacturer's instructions. The genomic DNA of the proband was subjected to whole exome sequencing (25). Illumina^® DNA Prep with Exome 2.5 Enrichment (cat. No. 20025524; Illumina, Inc.) was used to prepare DNA samples for whole exome sequencing. DNA was quantified using Qubit Fluorometric Quantitation (Thermo Fisher Scientific, Inc.). The quality of DNA was assessed by 1.5% agarose gel electrophoresis and visualized by ethidium bromide with using a GelDoc Go Gel Imaging System (Bio-Rad Laboratories, Inc.). The recommended read length of whole exome sequencing was 2x150 bp. SureSelect Human All Exon v6 (cat. No. 5190-8864, Agilent Technologies, Inc.) was used to capture the sequences. The sequencing library was prepared and the enriched library was subjected to whole exome sequencing on an Illumina HighSeq4000 (HiSeq 3000/4000 SBS Kit; cat. no. FC-410-1003; Illumina, Inc.). The loading concentration of the final sequencing library was 250-300 pM as determined using the Kapa Library quantification kit (part no. KK4824; Roche Diagnostics) and size-corrected using fragment analysis. Sequencing of both the forward and reverse strands was performed (paired end). After sequencing, the alignment of sequencing reads was performed using Burrows-Wheeler Aligner software (v0.59; https://bio-bwa.sourceforge.net/). Then, local realignment of the Burrows-Wheeler-aligned reads was performed by Genome Analysis Toolkit (GATK) IndelRealigner (26). Next, base quality recalibration of the Burrows-Wheeler-aligned reads was carried out using GATK BaseRecalibrator (26). Next, single-nucleotide variants (SNV) and insertions or deletions were identified using GATK Unified Genotyper (26). Then, these variants were annotated with the Consensus Coding Sequences Database (https://www.ncbi.nlm.nih.gov/CCDS/CcdsBrowse.cgi) at the National Centre for Biotechnology Information (NCBI). Illumina pipeline was used for image analysis and base calling. In addition, indexed primers were designed and used for data fidelity surveillance. SOAP aligner (soap2.21) software (27) was used to align the clean sequencing reads with human reference genome (hg19). Lastly, to assemble the consensus sequence and call genotypes in target regions, SOAPsnp (v1.05) software (28) was used.

Variants obtained from whole exome sequencing were collected. Variants with minor allele frequency (MAF) of <0.01 in dbSNP, HapMap, 1000 Genomes Project and our in-house database of 50,000 Chinese Han samples were selected. Public databases, namely, dbSNP (https://www.ncbi.nlm.nih.gov), HapMap (https://www.genome.gov), 1000 Genome Database (http://www.internationalgenome.org) and our in-house database for 50,000 Chinese Han samples were used (29-31). All of these databases were used to identify the MAFs of genetic variations in different populations. Using the MAF of a genetic variant in a specific population, the possible pathogenicity of this genetic variant for a specific case can be interpreted. The Human Gene Mutation database (HGMD, www.hgmd.cf.ac.uk/) contains all the reported genetic variants associated with monogenic disorders. Hence, by using this database, it can be interpreted whether the identified genetic variant is a novel or previously reported cause of any disease (32). Online Mendelian Inheritance in Man (OMIM, https://www.omim.org) is a highly comprehensive and freely available database, comprised of >16,000 human genes associated with all known Mendelian disorders (33). This database also provides information regarding genotype-phenotype correlation. Exome Aggregation Consortium (ExAC, http://exac.broadinstitute.org) contains exome sequencing data from largescale sequencing projects from different populations worldwide (34). Thus, the possible pathogenicity of identified genetic variants was confirmed by comparing their frequencies in different populations. Genome Aggregation Database (gnomAD, https://gnomad.broadinstitute.org) also comprises both genome and exome sequencing data from large-scale genome or exome sequencing projects from different populations around the world (35). Therefore, by comparing the frequencies of any reported genetic variant in different populations, the pathogenicity of the identified genetic variants in the present study was interpreted.

Variant interpretation was performed based on the variant interpretation guidelines of American College of Medical Genetics and Genomics (ACMG; Bethesda, Maryland, USA) (36). All the heterozygous, homozygous and compound heterozygous variants were selected based on gene function and disease association by OMIM and the literature. The expression of variants was confirmed using ‘Mutalyzer 2.0.35’ software (https://v2.mutalyzer.nl/) which strongly follows the Human Genome Variation Society guidelines (37). The whole exome sequencing quality control data are described in Table I.

Sanger sequencing was performed to validate the possible disease-causing mutations identified by whole exome sequencing. Primers for polymerase chain reaction (PCR) were designed based on the reference genomic sequences of the Human Genome from GenBank (NCBI; https://www.ncbi.nlm.nih.gov/genbank/). PCR products were subjected to Sanger sequencing. Sanger sequencing data were compared and analyzed.

The heterozygous novel mutation identified through whole exome sequencing was validated by Sanger sequencing. Sanger sequencing was performed with the following primers: Forward (F)'-5'-TGGTGGAGGGAAGGGATAGG-3'; reverse (R)5'-TGGCAAAGGGACAAGGGTTT-3'; the reference sequence NM_000070 of CAPN3 gene was used.

In order to understand the effect of the novel splice-site mutation of the CAPN3 gene on the splicing event of CAPN3 mRNA, RT-PCR followed by Sanger sequencing was performed. Muscle samples were collected from the proband and their family members, and total RNA was extracted using RNAqueous^™ Total RNA Isolation Kit (Thermo Fisher Scientific, Inc.) and reverse transcribed to cDNA using PrimeScript cDNA Synthesis Kit (cat. no. RR037A; Takara Biotechnology, Co., Ltd.) according to the manufacturer's recommendations. cDNA from the proband and their family members were amplified with the use of primers (F, 5'-CTGCTTCGTTAGGCTGGAGG-3'; R, 5'-GAAGCCTGTAGGGGTGTAGC-3') encompassing the coding sequence from exon 22 to exon 24. Then, the amplified cDNAs were run on a 2% agarose gel (2 g agarose powder was mixed with 100 ml 1xTAE in a microwavable flask and the gel was made). purified using DNAclear^™ Purification Kit (Thermo Fisher Scientific, Inc.). Finally, Sanger sequencing was performed as aforementioned for the proband and family members and the data were analyzed.

Reverse-transcribed cDNA was collected for fluorescence quantitative detection by quantitative (qPCR). The One-Step TB Green^® PrimeScript^™ RT-PCR Kit II (Perfect Real Time; cat. no. RR086A; Takara Biotechnology Co., Ltd.) was used. The thermocycling conditions were as follows: 15 min at 95˚C to activate the chemically modified hot-start Taq DNA polymerase, followed by 35-45 cycles of a 15-sec denaturation at 95˚C and 60 sec annealing and extension at 60˚C. The target gene and housekeeping gene (GAPDH; F-5'-GTCTCCTCTGACTTCAACAGCG-3' and R-5'-ACCACCCTGTTGCTGTAGCCAA-3') of each sample were subjected to (q)PCR. The change in RNA expression based on the detected Cq value was calculated. Primers used in qPCR were as follows: CAPN3 F, 5'-CTGCTTCGTTAGGCTGGAGG-3'; R, 5'-GAACTTACTTGAGGCATATG-3'. The RNA sample from the normal healthy individual was collected from our hospital's sample bank (The Reproductive Medicine Centre Sample Bank, The First Hospital of Lanzhou University, Lanzhou, China). The hospital's sample bank (The Reproductive Medicine Centre Sample Bank, The First Hospital of Lanzhou University, Lanzhou, China) collected the RNA sample from the normal healthy individual after getting written informed consent. Data were analysed using the comparative threshold cycle (2^-ΔΔCq) method (38).

The proband, a 16-year-old male patient from China, manifested with gradual and progressive weakness of upper and lower extremities. The clinical symptoms first appeared at 8 years old with slow, progressive weakness of muscles and an abnormal gait. At 10 years old, the proband presented with slow, progressive weakness of lower limbs with frequent falls, difficulty in standing, walking and climbing stairs. At the age of 16 years, physical testing revealed that the proband could only walk unaided for <15 min and that the proband was unable to raise their arms above their head. No sensory, ocular or bulbar abnormalities were identified, and neurological examination showed severe weakness of proximal muscles in all limbs, pelvic and shoulder girdles. At the age of 16 years, the proband had a normal mental status with no oculomotor or facial abnormalities with sensory and coordination examinations also finding no abnormalities. In the present study, bilateral atrophy of the biceps, shoulder muscles, hip adductors, posterior thigh muscles and knee flexors, as well as moderate hypertrophy on both sides of the calves and scapular winging, was reported. In order to assess muscle strength, the Medical Research Council (MRC) Scale was used (39). The MRC scale is divided into five grades (grade 5, normal muscle; grade 4, movement of the limb against gravity and resistance; grade 3, movement of the limb against gravity over the full range; grade 2, movement of the limb but not against gravity; grade 1, visible contraction without movement of the limb; grade 0, no visible contraction). There were six muscles examined (shoulder abductors, elbow flexors, wrist extensors, hip flexors, knee extensors and foot dorsiflexors) in both the upper and lower limbs on both left and right sides, each with a score from 0 to 5. Hence, the total score ranged from 60 (normal) to 0 (quadriplegic) (39). The muscle power of proximal upper and lower extremities was grade 3, while distal upper and lower extremities was 4+ according to the MRC scale. Moreover, deep tendon reflexes were normal in the proband's arm while no deep tendon reflexes were found in their lower extremities. Gowers' sign was used to ascertain the weakness of the muscle in the pelvic girdle or proximal muscle in lower limbs (40). This sign is a medical term that describes the strength of hip and thigh muscle, and is often identified amongst patients in the advanced stages of muscular dystrophies. The presence of Gowers' sign with decreasing plantar reflexes was identified in the proband. No abnormalities were found in the nerve conduction study; however, an electromyographic study revealed chronic myopathy. Thyroid tests were in the normal range; however, laboratory tests identified highly elevated levels (4,754 IU/l; normal, 35-232 IU/l) of serum creatine kinase. Normal levels of serum lactate and transaminases were observed in the proband. No abnormalities were identified in the electrocardiogram and echocardiogram, and the pulmonary function test (PFT) was normal.

According to these aforementioned clinical presentations, the proband was clinically diagnosed with LGMD.

The younger sister of the proband, a 14-year-old girl from China, manifested with progressive weakness of muscles since childhood. They presented with similar clinical symptoms, disease onset and progression as the proband. At 7 years old, they demonstrated difficulty standing, walking or running with frequent falls. These clinical symptoms gradually and progressively developed, and at 10 years old, they had exercise intolerance and difficulty standing from the sitting position independently. They developed weakness in both of the upper and lower extremities and finally lost the ability to ambulate independently at 11 years old. They also showed atrophy of both bilateral biceps and quadriceps with neurological examination showing weakness in the limb-girdle muscles. Their muscle strength was gradually and progressively reduced in the upper extremities of the proximal muscle, and lower extremities of both the proximal and distal muscles. Absence or diffusely reduced deep tendon reflexes were identified in the lower extremities; however, no abnormalities were found in extraocular movements and cognition. Diffuse atrophy in all upper and lower leg muscles was identified, PFT found no abnormalities, and physical and neurological examinations revealed that the muscle power of the proximal upper and lower extremities was grade 3 and 2, respectively, according to the MRC scale. The presence of Gowers' sign, as well as pelvic girdle and leg muscle atrophy, was observed; however, no abnormalities in facial muscles were identified. Laboratory tests found elevated levels (1,050 IU/l; normal, 35-232 IU/l) of serum creatine kinase; however, the levels of serum lactate and transaminases were normal. Electrocardiogram and echocardiogram showed no cardiac abnormalities, and no abnormalities were found in nerve conduction velocity.

The mother of the proband was a 42-year-old woman from China that presented with gradual, progressive weakness, and atrophy of both proximal and distal muscles. At 14 years old, they presented with difficulty running, climbing stairs, standing up from a sitting position and a waddling gait. At 18 years old, they struggled to walk independently and lift weights, and by their early twenties, they had lost the ability to walk independently, showing a complete loss of ambulation. They had an unremarkable medical history with normal sensation, no intellectual disabilities, and had never been reported to have fasciculations or pseudo-hypertrophy of calf muscles or joint contractures. However, neurological examination showed weakness and atrophy of limb-girdle muscles and an absence of all deep tendon reflexes. Laboratory tests found elevated levels (2,000 IU/l; normal, 35-232 IU/l) of serum creatine kinase. No abnormalities were identified in the levels of serum lactate and transaminases. The muscle power of both proximal upper and lower extremities, and distal upper and lower extremities were grade 2 according to the MRC scale, and the presence of Gowers' sign with atrophy of limb-girdle muscles was identified. No cardiac, motor or sensory nerve conduction abnormalities were identified; however, typical myopathic right biceps, severe atrophy of lower limbs and atrophy of paraspinal muscle were identified. At 30 years old, they developed back pain, hyperlordosis and myalgia, and both their proximal limb and abdominal wall muscles became weak. At present, they have manifested with paraspinal muscle atrophy.

MRI of the muscles of the proband was performed. The MRI revealed significant involvement of adductor magnus, thigh muscles (biceps femoris, semimembranosus and semitendinosus) and calf muscles (gastrocnemius and soleus) (Fig. 1B-E). Hypertrophy of the gracilis was detected; however, normal appearance of the rectus femoris, quadriceps femoris and sartorius of the thigh muscle was identified. Moreover, normal appearance of the tibialis anterior, tibialis posterior and extensor digitorum longus of the calf muscle was reported.

Clinical descriptions of the proband, their younger sister and mother have been comprehensively summarized in Table II.

Muscle biopsies of the proband showed mild to moderate dystrophic changes with H&E staining (Fig. 1F). These changes included increased fiber size variation with scattered atrophic and hypertrophic fibers, necrotic fibers undergoing myophagocytosis, grouped regeneration, endomysial fibrosis and fatty replacement (Fig. 1F). NADH staining revealed the mildly disorganized structure of the intermyofibrillar network (Fig. 1G) The immunostaining of caveolin-3, dystrophin, α-sarcoglycan and β-sarcoglycan indicated normal expression (positive staining) (Fig. 1H-K).

According to the aforementioned clinical symptoms and test results, the clinical diagnosis, therapeutic interventions and disease management were confirmed for the studied family.

Whole exome sequencing and Sanger sequencing identified a novel heterozygous splice-acceptor site mutation (c.2440-1G>A) in intron 23 of the CAPN3 gene in the proband. Sanger sequencing confirmed that this heterozygous novel mutation was also present in the mother and younger sister of the proband, while the father did not harbor the mutation (Fig. 1L).

The mutation was not present in 100 ethnically-matched normal healthy controls from our in-house database. This previously generated in-house control database collected genomic DNA samples from 100 normal healthy individuals, after getting their written informed consent, and performed whole exome sequencing to obtain raw sequencing data, followed by bioinformatics data analysis and interpretation. This normal control database has been used as a reference database, thus if any novel variant is identified in any gene, we can compare whether it is already present in the in-house database in order to understand its pathogenicity. This mutation was also absent in the HGMD, OMIM, ExAC, dbSNP, gnomAD and 1000 Genome Database, as well as our in-house database, which consists of ~50,000 Chinese Han samples.

The mutation (c.2440-1G>A) is classified as ‘likely pathogenic’ [absent in population database (PM2) + predicted null variant in a gene where loss-of-function is a known mechanism of disease (PVS1) + co-segregation of disease in multiple affected family members (PP1)] according to the variant interpretation guidelines of ACMG (36). The mutation (c.2440-1G>A) is co-segregated with the disease phenotype in this family with an autosomal dominant mode of inheritance.

Individuals with wild-type (father of the proband) and mutated (proband, their mother and younger sister) CAPN3 underwent PCR for amplification of CAPN3 cDNA encompassing the coding sequence (exon 22-24), which was run on a 2% agarose gel. A 200-bp band for wild-type CAPN3 cDNA was found, whereas a 176-bp band for mutated CAPN3 cDNA was identified (Fig. 2A).

Sanger sequencing of mutated cDNA from the proband, their mother and younger sister revealed that the mutation (c.2440-1G>A) at the last nucleotide of intron 23 of the CAPN3 gene disrupts the wild-type CAPN3 exon 24 splice acceptor-site and leads to aberrant splicing of CAPN3 mRNA, finally resulting in the skipping of exon 24 (24 bp). Complete loss of exon 24 causes the formation of a truncated CAPN3 protein (p.Trp814*) with the removal of eight amino acids from the C-terminal (Fig. 2B). However, Sanger sequencing of the wild-type CAPN3 cDNA showed normal splicing of exon 22 to exon 24 (Fig. 2B). The splicing mechanism of both wild-type and mutated cDNA is schematically presented (Fig. 2C).

Relative expression levels of CAPN3 mRNA revealed a significantly decreased mutated CAPN3 transcript level in the proband, their younger sister and mother while the father of the proband showed normal expression (same as the normal healthy control individual) of wild-type CAPN3 mRNA (Fig. 3). These results also suggested that the mutant CAPN3 transcript (proband, proband's mother and younger sister) was present at detectable levels. Therefore, the mutated CAPN3 transcript was not degraded by the nonsense-mediated mRNA decay pathway.

All 24 exons of the CAPN3 gene are schematically presented in Fig. 4A, and the mutation is located at the last nucleotide of exon 23. The wild type CAPN3 protein structure is shown in Fig. 4B. The mutated CAPN3 protein (with complete loss of exon 24) is presented in Fig. 4C. According to the structural point of view, this mutation causes partial loss of domain IV of the CAPN3 protein, which is involved in calcium binding and homodimerization (41).